Data envelopment analysis and game theory: Where are we? Where are we going?
摘要
This paper reviews the literature on using Data Envelopment Analysis (DEA) and Game Theory (GT). Given the importance of DEA in the performance measurement field and the fast growth of combined applications of the two techniques, we analyzed 180 publications indexed in the Scopus and Web of Science databases from 1980 to 2024. With the support of analytic tools, namely CitNetExplorer and VOSviewer, the main objective was to provide a bibliometric analysis to discuss the current state of the art, identify seminal articles, literature gaps, detect future trends, highlight the most applied researched areas and other techniques used in conjunction with DEA-GT. To classify the papers, we considered nine categories (type of study, research objectives, GT approach, game specificities, DEA model, DEA variables, application area and other techniques) that cover relevant perspectives of the field. The results indicated that 84% of the articles are applied papers. They use cooperative and non-cooperative game theory perspectives to improve the DEA results. The most widely applied DEA approaches are network, cross-efficiency, and classical models. Nash-bargaining games, Stackelberg games, zero-sum games, and imputation techniques must be highlighted in the GT framework. We identified the dominant application areas—banks and insurance companies, environment and energy—by analyzing the selected articles and their citations. Additionally, we identified and discussed 13 gaps, each one can be perceived as an opportunity for further research. Therefore, we provide guidelines in how to develop the field with future investigations.